Study On Stochastic Nature Of Gravel Bedload Transport

With the stratagem on the development of western China further carried out, in the research processes of development and protect of mountain rivers, the law of the flow and sediment movement in a gravel river is further becoming an important research task in the field of river dynamics. The thesis comments on the stochastic nature of gravel bedload motion and analyzes the stochastic nature by the field data. By undertaking experiments in a channel with complex boundary, the fluctuation feature of flow is analyzed. Further study on the statistical natures (e.g. probability distribution, turbulence intensity) of gravel bedload motion is undertaken. Especially, the effect of measured duration on turbulence intensity of bedload rate is discussed in detail. Applying fractal geometry, the stochastic process of gravel bedload rate is further explored. The main contents are as follows:By undertaking the experiments, the fluctuation of flow under the condition of complex boundary is investigated. The thesis explores the effect of the floodplain vegetation on the fluctuation of flow and obtains the distributions of the fluctuating velocity in streamwise, lateral and vertical directions. By analyzing the effect of the bed slope on the velocity, the velocity distributions for the cases of different bed slopes are obtained, In addition, the effect of the different types of vegetation on turbulence intensity is analyzed and the vertical distributions of turbulence intensities are obtained.Combined with the observed data, the stochastic nature of gravel bedloadmotion in Upper Yangtze River is investigated. The analysis results show that the deviation of bedload rate is very large, its statistical distribution scatters and gravel bedload motion bears the strong stochastic nature. Gravel motion shows intermittence, i.e. after traveling a short distance, gravel is at rest on the bed for a long time and its velocity is low. This nature results from deposition, sorting and abrasion of sediment that make sediment transport rate decrease along the channel. During the process of sediment transport, due to sorting in grain diameter, volume weight, and shape of sediment, gravel deposited cumulatively. Annual mean transport rates of gravel bedload for different stations are different and annual variation of transport rate of gravel for a given station is large. The shape of cross section of the channel and the features of the adjoining channel have distinct effects on the bedload rate.By the analysis of the field data of gravel bedload rate, it is found that the gravel bedload rate has a distinct stochastic nature and turbulence intensity of bedload rate is closely related to the measured duration. The deviation, diversity coefficient and dimensionless turbulence intensity and dimensionless ultimate difference of bedload rate per unit width, all decrease with the measured duration increasing.From the measured data, the frequency of the fluctuating bedlaod rate, f(qb), decreases with the increase of the bedload rate per unit width, q^ In general, as qb > 0.1 g/s.m, f{qb) ^ 1. The cumulative frequency of the gravel bedload rate per unit width follows P-III distribution. The maximum value of dimensionless ultimate difference of bedload rate per unit width is about 8.0. As the measured duration t > 600 s, the dimensionless ultimate difference Z < 4.0. As the measured duration t > 2400 s, the dimensionless ultimate difference Z < 2.0. The deviation coefficient of bedload rate per unit width has a maximum value of 2.0. As the measured duration t > 1200 s, the deviation coefficient Cv < 1.0. As the measured duration t > 3000 s, the deviation coefficient Cv < 0.5 . The dimensionless turbulence intensity of bedload rate per unit width has a maximum value of 10.0. As the measured duration /> 1200 s, the dimensionless turbulence intensity E'< 5.0. Asthe measured duration t > 3600 s, the dimensionless turbulence intensity E' < 2.0.Applying the modern mathematic tools such as fractal geometry, R/S analysis, the thesis analyzes the fractal feature of the income sediment from the upper Yangtze River. A positive correction is shown for sediment transport rate process in the upper Yangtze River. This means that the characteristics of the stochastic process are the features with long time memory and periods, by the computed value of Hurst exponent larger than 0.5. The research shows that fluctuation nature of the bedload rate may be an independent process or bear long time memory.